Naver Search MCP Server

Instructions

Implementation Reference

• server.py:458-471 (handler)

  The async function implementing the search_cafe_article tool logic, which calculates pagination parameters, prepares API params, and delegates the API call and response formatting to the shared _make_api_call helper.
  async def search_cafe_article(query: str, display: int = 10, page: int = 1, sort: str = "sim") -> str: """ Searches for cafe articles on Naver using the given keyword. The page parameter allows for page navigation and sort='sim'/'date' is supported. Args: query (str): The keyword to search for display (int, optional): The number of results to display. Default is 10. page (int, optional): The starting page number. Default is 1. sort (str, optional): The sorting criteria. Default is "sim" (similarity). """ start = calculate_start(page, display) display = min(display, 100) params = {"query": query, "display": display, "start": start, "sort": sort} return await _make_api_call("cafearticle.json", params, CafeArticleResult, "Cafe Article")
• server.py:454-457 (registration)

  The @mcp.tool decorator that registers the search_cafe_article tool with MCP, specifying its name and description.
  @mcp.tool( name="search_cafe_article", description="Searches for cafe articles on Naver using the given keyword. The page parameter allows for page navigation and sort='sim'/'date' is supported." )
• server.py:125-125 (schema)

  Output schema definition: Pydantic-style model for parsing the Naver Cafe Article search API response, inheriting from SearchResultBase and containing a list of CafeArticleItem objects.
  class CafeArticleResult(SearchResultBase): items: List[CafeArticleItem]
• server.py:58-60 (schema)

  Item schema for cafe articles: Extends DescriptionItem (which likely includes title, link, description) with cafe-specific fields cafename and cafeurl.
  class CafeArticleItem(DescriptionItem): cafename: Optional[str] = None cafeurl: Optional[str] = None